The proteolytic activity of matrix metalloproteases (MMPs)-2 and -9 is highly localized to the outer surfaces of invasive tumors. The goal of this project is to synthesize and evaluate prodrugs that are selectively cleaved by MMP-2 and -9 at the tumor site. Using both direct attachment and cleavable linkers, we will synthesize derivatives of two classes of antitumor compounds, the amino secocyclopropylbenzindolines (CBls) and doxorubicin, to which short MMP-cleavable peptides are appended. The CBIs are an extraordinarily potent class of minor groove binding/alkylating agents that show in vitro cytotoxicity at picomolar concentrations. Doxorubicin, white 1000-fold less potent in vitro, is a clinically approved drug having a well-described therapeutic profile against many hematologic and solid tumors. Our strategy takes advantage of the highly regulated, tumor-restricted proteotytic activity of MMPs-2 and -9, and the high turnover rates of these enzymes. We plan to demonstrate through our in vitro enzymatic assays, in vitro cytotoxicity assays, and in vivo therapy experiments, that MMP-substrate peptidyl prodrugs will have attenuated toxicities compared to their corresponding parent drugs, tumor selectivity, and therapeutic efficacy for the treatment of cancer. PROPOSED COMMERCIAL APPLICATION: Cancer therapy still represents one of the largest unmet medical needs. The goal of many prodrug strategies is to increase the amount of active drug delivered to a tumor through selective activation pathways. The prodrugs we propose are designed to be significantly less toxic until reactivated at the tumor site by tumor-selective enzymatic activity. this strategy could represent a significant advancement toward the goal of distinguishing between normal and malignant cells, and, as such, a significant advancement in the treatment of cancer.